Purdue University Graduate School
Browse

File(s) under embargo

Reason: Papers are either under review and waiting to be published

10

month(s)

10

day(s)

until file(s) become available

MUTUAL LEARNING ALGORITHMS IN MACHINE LEARNING

thesis
posted on 2023-05-18, 13:30 authored by Sabrina Tarin ChowdhurySabrina Tarin Chowdhury

    

Mutual learning algorithm is a machine learning algorithm where multiple machine learning algorithms learns from different sources and then share their knowledge among themselves so that all the agents can improve their classification and prediction accuracies simultaneously. Mutual learning algorithm can be an efficient mechanism for improving the machine learning and neural network efficiency in a multi-agent system. Usually, in knowledge distillation algorithms, a big network plays the role of a static teacher and passes the data to smaller networks, known as student networks, to improve the efficiency of the latter. In this thesis, it is showed that two small networks can dynamically and interchangeably play the changing roles of teacher and student to share their knowledge and hence, the efficiency of both the networks improve simultaneously. This type of dynamic learning mechanism can be very useful in mobile environment where there is resource constraint for training with big dataset. Data exchange in multi agent, teacher-student network system can lead to efficient learning.  

Funding

National Science Foundation under grant numbers 1930601 (to Yale) and 1930606 (to IUPUI).

History

Degree Type

  • Master of Science

Department

  • Computer Science

Campus location

  • Indianapolis

Advisor/Supervisor/Committee Chair

Snehasis Mukhopadhyay

Additional Committee Member 2

Shiaofen Fang

Additional Committee Member 3

Mihran Tuceryan

Usage metrics

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC